Short-term solar forecasting
Solar PV is rapidly becoming a significant source of humanity’s electricity. Fluctuations in solar PV output due to short-term events (like moving clouds) can have large impacts in areas with high solar PV penetration. This is particularly true where panels are geographically concentrated in industrial-scale PV farms. For this reason, significant effort has been made to forecast solar PV output using a variety of methods. Images of the sky contain a wealth of information, but this information is challenging to extract and use for reliable predictions. In the last 4 years, efforts have shifted to using machine vision systems to “read” the sky and make forecasts of PV panel output. Our group has developed a specialized convolutional neural network model named SUNSET (Stanford University Neural Network for Solar Electricity Trend) for 15 min ahead PV output forecast. The following four projects have been published, with more research projects going on.
Nowcast
We explore the inference of solar panel output solely from concurrent local sky images with a convolutional neural network (CNN). This research is the first time that CNNs – and by extension deep learning – have been applied to predict solar panel output. We demonstrate that sky images are useful in inferring PV panel output, and CNN is a suitable structure in this application.
Forecast
We propose a specialized convolutional neural network (CNN) “SUNSET” to predict 15-min ahead minutely-averaged PV output. The model is characterized by its usage of hybrid input, temporal history and strong regularization. Optimal input and output configurations are explored and suggestions are given. The code base of this work is available on Github.
Data fusion
We systematically explore 28 methods of “fusing” the heterogeneous inputs, i.e., PV power output history and ground-based sky images, in our CNN, to ensure that due importance is given to each type of input. We also systematically explore the many hyperparameters related to model training and tuning. Limited resources preclude an exhaustive search.
Sky condition specific sub-models
We propose a two-stage classification-prediction framework for the nowcast task and compare it with the end-to-end SUNSET model we developed in previous research. The proposed framework first classifies input images into different sky conditions and then the classified images are sent to specific sub-models for PV output prediction.
Publications
2020
Nie, Y., Sun, Y., Chen, Y., Orsini, R., & Brandt, A. (2020). PV power output prediction from sky images using convolutional neural network: The comparison of sky-condition-specific sub-models and an end-to-end model. Journal of Renewable and Sustainable Energy, 12(4), 046101. https://doi.org/10.1063/5.0014016
2019
Sun, Y., Venugopal, V., & Brandt, A. R. (2019). Short-term solar power forecast with deep learning: Exploring optimal input and output configuration. Solar Energy, 188, 730–741. https://doi.org/10.1016/j.solener.2019.06.041
Venugopal, V., Sun, Y., & Brandt, A. R. (2019). Short-term solar PV forecasting using computer vision: The search for optimal CNN architectures for incorporating sky images and PV generation history. Journal of Renewable and Sustainable Energy, 11(6), 066102. https://doi.org/10.1063/1.5122796
2018
Sun, Y., Szűcs, G., & Brandt, A. R. (2018). Solar PV output prediction from video streams using convolutional neural networks. Energy & Environmental Science, 11(7), 1811–1818. https://doi.org/10.1039/C7EE03420B
Site content
- Sherwin, Evan, Jeffrey Rutherford, Yuanlei Chen, Sam Aminfard, Eric Kort, Robert Jackson, and Adam Brandt. “Single-Blind Validation of Space-Based Point-Source Detection and Quantification of Onshore Methane Emissions”, Scientific Reports, 13 (March 7, 2023): 3836. https://doi.org/10.1038/s41598-023-30761-2.
- Jing, Liang, Hassan El-Houjeiri, Jean-Christophe Monfort, James Littlefield, Amjaad Al-Qahtani, Yash Dixit, Raymond Speth, Adam Brandt, Mohammad Masnadi, Heather MacLean, William Peltier, Deborah Gordon, and Joule Bergerson. “Understanding Variability in Petroleum Jet Fuel Life Cycle Greenhouse Gas Emissions to Inform Aviation Decarbonization”, Nature Communications, 13, no. 1 (December 21, 2022): 7853. https://doi.org/10.1038/s41467-022-35392-1.
- Zhang, Zhan, Evan Sherwin, Daniel Varon, and Adam Brandt. “Detecting and Quantifying Methane Emissions from Oil and Gas Production: Algorithm Development With Ground-Truth Calibration Based on Sentinel-2 Satellite Imagery”, Atmospheric Measurement Techniques, 15, no. 23 (December 13, 2022): 7155-69. https://doi.org/10.5194/amt-15-7155-2022.
- Sherwin, Evand, Ernest Lever, and Adam Brandt. “Low-Cost Representative Sampling for a Natural Gas Distribution System in Transition”, ACS Omega, 7, no. 48 (November 23, 2022): 43973–43980. https://doi.org/10.1021/acsomega.2c05314.
- Yu, Jevan, Benjamin Hmiel, David Lyon, Jack Warren, Daniel Cusworth, Riley Duren, Yuanlei Chen, Erin Murphy, and Adam Brandt. “Methane Emissions from Natural Gas Gathering Pipelines in the Permian Basin”, Environmental Science & Technology Letters, 9, no. 11 (October 4, 2022): 969–974. https://doi.org/10.1021/acs.estlett.2c00380.
- Kuepper, Lucas, Holger Teichgraeber, Nils Baumgärtner, André Bardow, and Adam Brandt. “Wind Data Introduce Error in Time-Series Reduction for Capacity Expansion Modelling”, Energy, 256 (October 1, 2022): 124467. https://doi.org/10.1016/j.energy.2022.124467.
- Plant, Genevieve, Eric Kort, Adam Brandt, Yuanlei Chen, Graham Fordice, Alan Gorchov Negron, Stefan Schwietzke, Mackenzie Smith, and Daniel Zavala-Araiza. “Inefficient and Unlit Natural Gas Flares Both Emit Large Quantities of Methane”, Science, Report: Methane Emissions, 377, no. 6614 (September 29, 2022): 1566-71. https://doi.org/10.1126/science.abq0385.
- Von Wald, Gregory, Kaarthik Sundbar, Evan Sherwin, Anatoly Zlotnik, and Adam Brandt. “Optimal Gas-Electric Energy System Decarbonization Planning”, Advances in Applied Energy, 6 (June 2022): 100086. https://doi.org/10.1016/j.adapen.2022.100086.
- Teichgraeber, Holger, and Adam Brandt. “Time-Series Aggregation for the Optimization of Energy Systems: Goals, Challenges, Approaches, and Opportunities”, Renewable and Sustainable Energy Reviews, 157 (April 2022): 111984. https://doi.org/10.1016/j.rser.2021.111984.
- Chen, Yuanlei, Evan Sherwin, Elena Berman, Brian Jones, Matthew Gordo, Erin Wetherly, Erik Kort, and Adam Brandt. “Quantifying Regional Methane Emissions in the New Mexico Permian Basin With a Comprehensive Aerial Survey”, Environmental Science & Technology, 56, no. 7 (March 23, 2022): 4317–4323. https://doi.org/10.1021/acs.est.1c06458.
- Shi, Lin, Katharine Mach, Sangwon Suh, and Adam Brandt. “Functionality-Based Life Cycle Assessment Framework: An Information and Communication Technologies (ICT) Product Case Study”, Journal of Industrial Ecology, 26, no. 3 (February 15, 2022): 782-800. https://doi.org/10.1111/jiec.13240 Find it @ Stanford.
- Bell, Clay, Jeff Rutherford, Adam Brandt, Evan Sherwin, Timothy Vaughn, and Daniel Zimmerle. “Single-Blind Determination of Methane Detection Limits and Quantification Accuracy Using Aircraft-Based LiDAR”, Elementa: Science of the Anthropocene, 10, no. 1 (January 4, 2022): 00080. https://doi.org/10.1525/elementa.2022.00080.
- Wang, Jinfang, Jingwei Ji, Arvind Ravikumara, Silvio Savarese, and Adam Brandt. “VideoGasNet: Deep Learning for Natural Gas Methane Leak Classification Using an Infrared Camera” 238 (January 1, 2022): 121516. https://doi.org/10.1016/j.energy.2021.121516.
- Teichgraeber, Holger, Lucas Küpper, and Adam Brandt. “Designing Reliable Future Energy Systems by Iteratively Including Extreme Periods in Time-Series Aggregation”, Applied Energy, 304 (December 15, 2021): 117696. https://doi.org/10.1016/j.apenergy.2021.117696.
- Zhang, Zhan, Evan Sherwin, and Adam Brandt. “Estimating Global Oilfield-Specific Flaring With Uncertainty Using a Detailed Geographic Database of Oil and Gas Fields”, Environmental Research Letters, 16, no. 12 (November 30, 2021): 124039. https://doi.org/10.1088/1748-9326/ac3956.
- El Abbadi, Shar, Evan Sherwin, Adam Brandt, Stephen Luby, and Craig Criddle. “Displacing Fishmeal With Protein Derived from Stranded Methane”, Nature Sustainability, 5 (November 22, 2021): 47–56. https://doi.org/10.1038/s41893-021-00796-2.
- Masnadi, Mohammad, Giacomo Benini, Hassan El-Houjeiri, Alice Milivinti, James Anderson, Timothy Wallington, Robert De Kleine, Valerio Dotti, Patrick Jochem, and Adam Brandt. “Carbon Implications of Marginal Oils from Market-Derived Demand Shocks”, Nature Research, 599, no. 7883 (November 3, 2021): 80–84. https://doi.org/10.1038/s41586-021-03932-2.
- Orsini, Rachel, Philip Brodrick, Adam Brandt, and Louis Durlofsky. “Computational Optimization of Solar Thermal Generation With Energy Storage”, Sustainable Energy Technologies and Assessments, 47, no. 3 (October 2021): 101342. https://doi.org/10.1016/j.seta.2021.101342.
- Rutherford, Jeffrey, Evan Sherwin, Arvind Ravikumar, Gavin Heath, Jacob Englander, Daniel Cooley, Mark Omara, Quinn Lanfitt, and Adam Brandt. “Closing the Methane Gap in US Oil and Natural Gas Production Emissions Inventories”, Nature Communications, 12, no. 1 (August 5, 2021): 4715. https://doi.org/10.1038/s41467-021-25017-4.
- Nie, Yuhao, Ahmed Zamzam, and Adam Brandt. “Resampling and Data Augmentation for Short-Term PV Output Prediction Based on an Imbalanced Sky Images Dataset Using Convolutional Neural Networks”, Solar Energy, 224 (August 2021): 341-54. https://doi.org/10.1016/j.solener.2021.05.095.